Open Chinese LLM - GSM8K — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 338 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2 72B Instruct | 78.7 |
| 2 | Qwen 2 72B | 75.36 |
| 3 | Meta-LLama-3-Cat-Smaug-LLama-70B | 72.93 |
| 4 | Llama3-TenyxChat-70B | 72.71 |
| 5 | Higgs-Llama-3-70B | 71.65 |
| 6 | Le_Triomphant-ECE-TW3 | 71.57 |
| 7 | llama-3-70B-Instruct-abliterated | 71.34 |
| 8 | dolphin-2.9.1-llama-3-70B | 70.96 |
| 9 | Llama 3 70B Japanese Suzume Vector V0.1 | 70.58 |
| 10 | TW3-JRGL-v2 | 70.36 |
| 11 | YiSM-34B-0rn | 70.36 |
| 12 | free-evo-qwen72B-v0.8-re | 70.28 |
| 13 | Rhea-72B-v0.5 | 70.2 |
| 14 | Smaug-72B-v0.1 | 70.13 |
| 15 | YiSM-blossom5.1-34B-SLERP | 70.13 |
Interactive version: theaggregate.ai/benchmark?slug=open-chinese-llm-gsm8k · How the rankings work · Data refreshed daily, snapshot 2026-07-22.